Iron overdose epidemiology, clinical features and iron concentration-effect relationships: the UK experience 2008-2017.

Iron poisoning is potentially serious, but mortality has fallen worldwide since implementation of pack size and packaging restrictions, and changes in iron use during pregnancy. The management of individual cases of overdose remains problematic due to uncertainty about indications for antidote. We examine the epidemiology of iron overdose in hospital cases referred to the UK National Poisons Information Service (NPIS) and evaluate the toxicokinetics of iron in patients ingesting only iron preparations. Anonymized hospital referral patient data from the NPIS database were collated for the period 1 January 2008 to 31 July 2017. Information was extracted, where recorded, on type of ingestion [iron alone (single), or combined with other agents (mixed)], reported dose, iron salt, timed iron concentrations and symptoms. In single-agent ingestions, the relationships between reported elemental iron dose, early concentrations (4-6 h), and symptoms were evaluated in teenagers and adults (≥13 years) and children (≤12 years) using standard statistical techniques (correlation and unpaired nonparametric comparisons). In those patients with sufficient sample points (three or more), a simple kinetic analysis was conducted. Of 2708 patients with iron overdoses referred by UK hospitals for advice during the 9.7 years study period, 1839 were single-agent ingestions. There were two peaks in age incidence in single-agent exposures; 539/1839 (28.4%) were <6 years (54.1% males) while 675/1839 (36.7%) were between 13 and 20 years (91% females), the latter a substantial excess over the proportion in the totality of hospital referrals to the NPIS in the same period (13-20 years: 23,776/144,268 16.5%; 67.5% female) ( < .0001 overall and for female %). In 475 teenagers and adults and 86 children, with at least one-timed iron concentration available, there was no correlation between stated dose and iron concentration measured 4-6 h post-ingestion. Observed peak iron concentrations were not related to reported symptoms in adults. Initial iron concentrations were significantly higher in 30 patients (25 adults, 5 children) who received desferrioxamine (DFO) compared to those that did not [no DFO: mean 63.8 μmol/L (95% CI 62.1-65.6), median 64; DFO: mean 78.5 μmol/L (95% CI 69.2-87.7), median 78.1; Mann-Whitney  < .0018). No significant differences in symptoms were observed pre-treatment between DFO-treated and untreated groups. No patients died in this cohort. Single-agent iron exposures reported from UK hospitals were most common in children <5 years and young people aged 13-20 years. Poisoning with organ failure was not identified and there were no fatalities. No correlations were observed between reported iron doses and early concentrations, or between iron concentrations and symptoms in this cohort of mild-to-moderate poisoning.